Engaging Young Men as Community Health Leaders in an STI and Intimate Partner Violence Prevention Trial in Dar es Salaam, Tanzania

Abstract

Background:

This article presents lessons learned from a microfinance and health intervention for young men designed to prevent sexually transmitted infections (STI) and intimate partner violence (IPV) in Dar es Salaam, Tanzania. We describe the different strategies we used to identify and train young men to become change agents within their social networks.

Description:

A cluster-randomized trial with 60 camps was undertaken in the Kinondoni District of Dar es Salaam. A total of 170 members from 30 intervention camps were trained in March 2014 as popular opinion leaders (POLs), whom we call Camp Health Leaders (CHLs). We describe the process of nominating, training, and retaining CHLs. We also describe our monitoring process, which included the collection of weekly diaries assessing topics discussed, number of peers engaged in conversations, reactions of peers, and challenges faced.

Lessons learned:

POLs within naturally existing social networks can be engaged in STI and IPV prevention initiatives. Continuous efforts in retention, such as holding community advisory board (CAB) meetings, developing prevention slogans and t-shirts, and offering small grants to POL teams for intervention activities, were important to keeping POLs engaged in the intervention. Further, booster-training sessions were critical to maintain the message of the project and ensure that the challenges POLs face with implementing the programs were addressed in a timely manner.

Conclusion:

Recruiting POLs in a Tanzanian urban setting and engaging them in STI and IPV reduction through social networks is possible. Training POLs in health information and interpersonal communication is important. Utilizing booster sessions and a variety of retention strategies for POLs in programs that aim to reduce IPV and STI infections among young men is essential to maintain the health leaders’ engagement in the intervention as well as intervention fidelity.

INTRODUCTION

For most Tanzanian youth, transitioning to adulthood has significant social, health and economic risks. The HIV pandemic in sub-Saharan African settings such as Tanzania has disproportionately affected youth. Worldwide, more than half of those infected with HIV are between 15–24 years of age (UNAIDS, 2010). About 5.1% of Tanzanians aged 15–49 are currently living with HIV (NBS, 2013). Further, 1% of Tanzanian adolescents aged 15–19 years are living with HIV and there is a sharp increase in infection rates for those aged 20–24 reaching 3.2% (TACAIDS, 2013). Therefore, intervening with youth is vital in stemming the trajectory of HIV infection.

Gender based violence and specifically intimate partner violence (IPV) is commonly reported by women in Tanzania as one of the multitude of hazards they face. In 2005, among women in Dar es Salaam, the commercial capital, 41.3% had ever experienced physical or sexual violence (33% physical, 23% sexual 23%); prevalence within the Tanzanian region of Mbeya were higher at 56% (47% physical, 31% sexual)(WHO, 2005). More recently, 65% of ever-married or partnered women in Mwanza, the second most populous city, reported ever experiencing IPV (34% emotional, 18% physical, 21% and sexual) (Ezekiel & Chitama, 2016). Further, recent evidence in Tanzania has shown that while IPV perpetration is much higher among men; it may also be bidirectional within relationships.

IPV victimization is linked to an increased risk of STIs including HIV (WHO, 2004; Jewkes et al., 2006; Li et al., 2014). Numerous HIV risk behaviors, including IPV, are precipitated by gender norms. Ideas of masculinity and manhood have long been recognized as drivers of violence and unsafe sexual behavior (Rivers & Aggleton, 1998; UNAIDS, 2000). In addition to IPV, masculine gender role ideologies and inequitable gender norms have been associated with other HIV risky behaviors including concurrency (Eaton & Case, 2009; ) and unprotected sex (Noar & Morokof, 2001; Santana et al., 2006; Pulerwitz & Barker, 2008). For over 20 years gender parity has been recognized as a prerequisite to effectively halt the spread of HIV (Larson, 1996). Calls to engage men in behavior change related to sexual behavior and IPV have resulted in numerous projects, including some that have demonstrated to be effective (Barker et al., 2010). Young men remain vital in the fight against STIs and IPV.

Engaging young men as community health leaders

Social influence approaches to behavior change, including the popular opinion leader (POL) models, have been widely utilized in understanding and addressing population risk behavior patterns (Kelly et al., 1991). POLs, well known and often respected individuals within their social networks, are increasingly commonly recruited as change agents within their networks to disseminate preventative health messaging as part of interventions. Based on the theory of diffusion of innovations (Rogers, 2003), the POL model aims to affect the attitudes, beliefs, and behaviors of others by engaging opinion leaders who have social status and influence among their peers (Valente & Pumpuang, 2007). POLs may act as intermediaries or partners of the intervention team, communicating behavior change messages through various channels such as word of mouth, social media and text messages (Ko et al., 2013; Theall, Fleckman, & Jacobs, 2015; Young et al., 2015; Odeny et al., 2017), serving as models for behavior change (Valente & Pumpuang, 2007), and helping their peers evaluate and make the decision to adopt the recommended behavior change (Rogers, 2003).

POLs engaged in HIV prevention activities have successfully reduced stigma and increased testing among Peruvian men who have sex with men (MSM) (Young et al., 2011; Young et al., 2015), decreased stigma and increased communication on HIV messages in China (Rice, Wu., Li, Detels, & Rotheram-Borus., 2012), increased HIV testing and consistent condom use among MSM in Taiwan (Ko et al., 2013), reduced the number of sexual partners among African American men, and increased use of condoms among gay Latino youth (Somerville, Diaz, Davis, Coleman, & Taveras., 2006). In contrast to these peer reviewed POL-based HIV prevention studies which demonstrated success in behavior change, one study in London showed no impact on HIV risk behaviors and another in Scotland did not show community-level effects (Elford, Bolding, & Sherr, 2004; Flowers, Hart, Williamson, Frankis, & Der, 2002). However, another author suggested that these studies did not adhere to all core elements of POL engagement, specifically, these programs were noted to fall short in critical mass of POLs engaged, and depth of training. In response, authors of the London study claimed all core elements were engaged, but POL recruitment and language barriers proved to be obstacles to implementation (Elford, Bolding, & Sherr, 2004). Successful engagement of POLs may require a combination of full implementation, POL uptake of messages and strategies, as well as adaptation to challenges encountered in the field.

Because of the normative nature of sexual risk and IPV in young men’s peer groups, we developed an intervention to leverage the social influence of young men’s peers based upon the POL model of social influence to prevent HIV/STI transmission and intimate partner violence in social groups of young men known as “camps” in Dar es Salaam, Tanzania. Camps are stable, organized social networks of mostly young men that meet in fixed neighborhood locations for business or social activities, have constitutions, elected leadership and membership. In the context of the intervention we called POLs “camp health leaders” (CHLs).

Overview of the intervention

We implemented our intervention within camps in four wards of Dar es Salaam. Camps were identified through a formative study designed to identify venues where young men engaging in high-risk behaviors socialize. The intervention consisted of a combined microfinance and health leadership approach with the aim of reducing STI incidence and the perpetration of IPV by addressing the gender norms contributing to increased risk of STI acquisition and IPV. The health leadership intervention was designed to transform men’s attitudes towards gender roles to be more equitable by changing camp-level norms. The intervention focused on addressing norms and behaviors that lead to HIV risk and intimate partner violence. For example, the training addressed male gender norms wherein men have power and control to determine when to have sex and whether to use protection or not. The training helped CHL develop messages to address these gender norms. Through formative work we found engaging young men to disseminate STI and IPV prevention messages feasible. This formative pilot study was conducted within four camps within one ward of Dar es Salaam (Tandale) and the current study scaled this intervention to reach 30 camps across four wards (Mabibo, Manzese, Mwananyamala and Tandale). A three step, probability based sampling method was used to randomly select 60 eligible camps into the trial, which were then divided into treatment (n=30) and control (n=30). We also increased time of engagement from six months for the pilot to two years for the trial. The purpose of this paper is to summarize the major lessons learned in the scaled-up implementation of the CHL component of this intervention. The main research questions were: 1) How can we recruit CHLs 2) How can we train and assess CHLs’ training performance and 3) How can we retain CHLs. Specifically, we will focus on the core strategies we used in each of these areas for the scaled CHL engagement.

METHODS

Recruitment

We used the PLACE (Priorities for Local AIDS Control Efforts) method to identify camps in four wards of Kinondoni District in Dar es Salaam Tanzania (Weir et al., 2005). Camps were identified through community informant interviews throughout these wards, and we confirmed their existence and operation via camp verification interviews. A total of 205 camps were located and operational, 179 of which met our eligibility criteria, and 60 selected in to our study, and one was later found to be ineligible. We then convened a meeting with the camp leaders at a local government’s office, introduced our study and shared our study goals. We then shared with the camp leaders our plans of working with the camps, starting with the CHL selection. Later, meetings were held with camp members during which members nominated their peers whom they considered to have leadership qualities. Meetings were held in each camp, during which members brainstormed and listed leadership qualities. When the camp members did not mention leadership qualities listed in our standard operating procedures that the team considered important, our intervention coordinator added them to the ongoing list at the meeting. Camp members then voted for members they considered to have most of the mentioned qualities by writing three names on a piece of paper. The top 20% of the total number of camp members with highest nominations in each camp were selected as CHLs.

Selected CHLs had to have completed a baseline survey and attend a training to be eligible. 170 CHLs were selected from a total of 185 who were nominated initially, of which 10 did not attend training, 3 had falsified their identity and were not in registered as camp members, and 2 had not completed baseline survey and were therefore excluded. The majority (64.7%) of the CHLs were 20–29 years old and 34% had completed secondary school (Table 1). While 28% were married and more than half (58%) had been tested for HIV. The selected CHLs were informed about the project’s focus of enabling young men to protect themselves from STIs and reduce IPV, and that the CHLs were a vessel for the project to endorse safer and protective sexual behaviors and IPV reduction in their respective camps. The selected CHLs were also informed that they would be trained on ways to reduce risky sexual behaviors that may lead to acquiring STIs, to address gender norms to reduce IPV, and to use effective communication. CHLs were also made aware that after training, they were expected to incorporate these lessons into casual conversations with the peers in their networks.

Table 1.

Demographic characteristics and HIV testing behavior of camp health leaders

	n = 170	%
Age
15–19	23	13.5
20–24	58	34.1
25–29	52	30.6
30+	37	21.8
Education
Primary school or less	92	54.1
Some secondary school	21	12.4
Secondary school completed	57	33.5
Marital status
No	121	71.6
Yes	48	28.4
Previous HIV test
No	70	41.2
Yes	100	58.8

Training

Nominees were required to attend a five-day initial training and one-day booster sessions every six months for a period of two years. Attendance was taken daily during the initial training and nominees who missed a training (n=10) did not become CHLs. To maintain high levels of attendance, the study team contacted camp leaders daily to remind CHLs to attend the next day. The training comprised of addressing STI myths, STI prevention including correct use of condoms, and the associated risks of IPV and multiple concurrent sexual partnerships. The IPV reduction content included understanding IPV and norms around IPV. The training module on IPV used several methods adapted from the SASA! Toolkit (Michau et al., 2008), which has been used previously in East Africa. These included several interactive activities such as “In Her Shoes” and “My Planet” that were designed to help the peer leaders clarify their own attitudes and values related to gender, violence, and power. These activities also aimed to help the CHLs understand how IPV affects women’s health and well-being. Skill building using participatory methods was a major component of the training. Throughout the program, CHLs also learned communication strategies including how to counter negative viewpoints and participated in role-playing and demonstrations whereby they learned different strategies to engage their peers in conversations and how to improve their listening skills. After the training, CHLs were asked to implement the strategies they learned in the camps by incorporating what they learned in naturally occurring conversations with peers.

Evaluation

To evaluate the effectiveness of the training, we administered a pre-and post-training questionnaire-using pen and paper. The pre-training questionnaire was administered on the first day of the training, prior to the first session and the post-training questionnaire was administered on day five, after the last session. Topics assessed on these questionnaires included general knowledge about HIV (measured with a series of 9 yes/no questions) and confidence level to talk with camp members about different HIV and IPV related topics (measuring using a 4-point Likert-type scale ranging from 1 = very unconfident to 4 = very confident).

To evaluate the ongoing process of CHLs implementing the health leadership intervention, the team provided diaries that the CHLs were asked to complete weekly, recording the different conversations about IPV, condom use or STI/HIV they engaged in with peers. A tracer, who was part of the study team, visited the camps to collect the diaries each month and also to assist CHLs who experienced challenges in filling out the diaries. A data entrant then typed the diaries while also noting any concerns that needed the team’s immediate response

Data analysis

First, univariate analyses were conducted to examine the distribution of the sociodemographic and previous HIV testing of CHLs. Second, an evaluation of the training CHLs received to assess their level of HIV/AIDS knowledge and their confidence level to talk with camp members about HIV and IPV was conducted by analyzing the pre-and post-training questionnaires. For these bivariate analyses, we combined “very unconfident” and “somewhat unconfident” as “unconfident,” and “very confident” and “somewhat confident” as “confident”. Complete pre-and-post training data for all 170 CHLs were included in the analyses. To adjust for the multiple observations within the randomized clusters, null hypotheses of no change pre to post were tested using Rao-Scott’s chi-square. SAS statistical software version 9.3 (SAS Institute Inc., Cary, NC) was used to conduct the analysis.

Retention

Three retention strategies were used to keep CHLs engaged and motivated with the program. The first retention strategy was that the team utilized was booster training sessions. These sessions were held every six months and focused on reviewing knowledge and skills that were acquired during the initial training. The booster training sessions were also utilized to review successes and challenges that the CHLs faced in implementing the planned strategies. CHL who were noted to be late to complete diaries, or who suggested that they faced specific challenges were given technical assistance during the booster sessions. A modest allowance was given to the attendees to cover transportation costs getting to and from the training venue. Attendance registers were used to record CHL attendance.

The second retention strategy utilized was community advisory board (CAB) meetings. These were also held every six months in each of the four study wards. Members who were part of the CAB included camp leaders from all intervention and control camps, community (ward) leaders as well as one camp guardian from each camp. In these meetings, the CAB was given an update of what has been happening in the previous six months and also future plans. Attendance registers were also used to record attendance of all invitees.

The third retention strategy included providing CHLs in each camp with small grants (approximately US $150) to support health related activities such as IPV and STI awareness music concerts, sports bonanzas that included a brief presentation by the CHLs, and workshops where they discussed IPV and STI related topics with peers. To apply for the grant, CHLs had to create an idea, discuss it with the study team, and receive suggestions from the study team for improvements before the idea was approved. Importantly, CHLs were also involved in the worldwide reduction campaign known as the “16 days of activism.” In this campaign that is held every year, the CHLs were briefed on the message for the year during the booster sessions and were asked to include the message in their conversations for the next sixteen days. All camps received these funds and conducted different activities such as soccer and netball matches, music – including hip hop with IPV a or STI messages.

Institutional Review Board approval was applied for and acquired from.

RESULTS

Lessons learned and conclusions

Recruitment

Engaging young male POLs in STI and IPV prevention is possible

This study demonstrated that engaging young male popular opinion leaders in a scaled-up STI and IPV prevention intervention was feasible. Camp members nominated 185 of their peers to be CHLs, and 170 completed the training and were engaged as CHLs. With sufficient notice and reminders to their camp leaders, these young men successfully attended a required five-day training as well as the booster sessions. Further, the CHLs’ engagement with the study is shown by the fact that they initiated the camp activities that the project funded through the small grants.

Training

HIV/AIDS Knowledge and Confidence Level Assessment

A total of 170 CHLs from 30 camps were trained and successfully engaged as CHLs. Table 2 shows the proportion of CHLs who answered yes for questions assessing their HIV transmission knowledge pre-and–post training. While most CHLs were knowledgeable about HIV prior to the training, there was a significant increase post-training in the proportion of CHLs who reported that a person can get HIV by: 1) sharing needles to inject drugs (95.5% vs. 99.4%, p = <0.01); 2) having unprotected anal sex (88.1% vs. 96.5%, p = <0.001); and 3) having oral sex (74% vs. 90%, p = < 0.001). In addition, there was a significant increase in the proportion of CHLs who reported that when used correctly, condoms can protect people from becoming infected with HIV (89.9% vs. 97.7%, p = <0.01), and that HIV is transmitted more easily during dry sex than wet sex (84% vs. 93.5%, p = <0.0.01).

Table 2.

Camp health leaders’ pre-and-post training HIV knowledge assessment

Questions	Pre-training Yes (%)	Post-training Yes (%)	p-values
A person can get HIV by sharing needles to inject drugs	95.5	99.4	<0.01
A person can get HIV by sharing food or clothing with a person living with HIV	6.1	2.3	<0.05
A person can get HIV by having unprotected anal sex	88.1	96.5	<0.001
A person can get HIV by getting a blood test	12.9	12.2	0.58
A person can get HIV by having oral sex	74	90	<0.001
When used correctly, condoms can protect people from becoming infected with HIV	89.9	97.7	<0.01
HIV is transmitted more easily during dry sex than wet sex	84	93.5	< 0.01
A person who looks physically strong and healthy can be HIV positive	83.1	85.1	0.72

As shown in Table 3, there was an increase in the proportion of CHLs who reported feeling confident to talk with camp members about the different topics but the increase was only significant for challenging the idea that men have the power to decide when to have sex (69.9% vs. 75.2%, p <0.05).

Table 3.

Camp health leaders’ confidence to talk with camp members about HIV and GBV

Topics	Pre-training confidence (%)	Post-training confidence (%)	p-values
Using condoms correctly and consistently	85.4	94.7	0.21
Strategies in negotiating condom use with sexual partners	87.4	93.5	0.53
The risk of having multiple sexual partners at one time	61	71	0.26
The impact of alcohol on sexual health	64.7	67.9	0.83
The association of gender based violence and HIV	64.9	78	0.84
The importance of HIV testing	90.3	94.6	0.17
Male has a power to decide when to have sex	69.9	75.2	<0.05
How to use non-violent strategies to resolve conflict	91	95.8	0.29

Retention

Retention in the program was measured by the filling and submission of diaries. In the two years of diary submissions (a total of 104 weeks), all but one CHL submitted at least one diary, 11 CHL submitted 76% or more weeks, 64 CHL submitted 51–75% of weeks, 80 submitted 26–50% of weeks, 4 submitted 25% or less of weeks. CHL diary entries showed trends where they seemed to be mostly talking about HIV than IPV. And within HIV, talking about condoms was recorded more frequently than concurrency or HIV testing.

Though most CHLs reported feeling confident after the training, they encountered several challenges that they reported during the booster sessions. One challenge that was first mentioned was that the diaries were too comprehensive and required too much of their time; after which they were reduced in the number of pages and also were collected once a month. Few of the CHLs seemed to have an issue with filling out the diaries, due to various reasons including inability to read and write whereby the tracer for the team helped them complete the diaries by discussing with them the conversations they had had and helping them to fill the diaries.

Another challenge reported by some of CHLs was difficulty in initiating conversations because fellow camp members did not take them seriously in their “new” role. The intervention team addressed these challenges during the booster sessions, where CHLs could also hear and learn about the successful strategies other CHLs were using in their respective camps. Some CHLs however, acknowledged that it was easier for their peers to listen and change when the CHLs stopped engaging in sexual risky behaviors. These sessions helped to motivate CHLs who reported challenges initiating conversations for other reasons such as being shy. The study team reminded CHLs that they did not need to teach, but rather to engage their peers in naturally occurring conversations. In addition, we also printed t-shirts with HIV and IPV slogans for CHLs to wear as conversation starters. These slogans were created by CHLs themselves in each CHL class and then they voted for the best slogan. A total of six slogans were selected from the six CHL classes and then a final vote for the six were held where the one that won was “Naacha Ukatili wa Kijinsia, Wewe?” in Swahili, which translates to “I’m stopping GBV, You?” in English. Without the comprehensive, participatory skill building training, and additional support from the research team, it would not have been possible for CHLs to hold weekly conversations with peers.

CONCLUSIONS

Recruiting CHLs through camps is possible given the stability of the camps. Our study showed that scaling this approach to recruit young men as POLs in STI and IPV reduction activities can be done if one engages the camp leaders from inception with a variety of strategies and is responsive to health leader suggestions. Creativity in retention strategies is vital. The program utilized different retention strategies that improved the CHLs’ participation and engaged their interests over two years. Our team worked with the CHLs by using their own ideas in improving the program such as the camp health activities that the CHLs developed. This ensured that the CHLs felt valued and that they owned the program. The small grants idea initially came from CHLs from a few camps, who had asked for funding to do some awareness raising for fellow camp members who had not attended the training. Further, we also show that these CHLs from camps can be trained in various methods that would enable them to carry out STI and IPV reduction conversations in naturally occurring conversations with their peers. Finally, it was also important to acknowledge CHLs’ commitment to STI and IPV reduction by engaging them in national and international activities such as the “16 days of activism.” Future studies focusing on engaging men in GBV and STI reduction using POLs should consider increasing the percentage of the POLs from networks as well as including observational methods regarding CHL performance.

Limitations

Several limitations are associated with the engagement of CHLs. The first is the use of diaries, which was self-reported and therefore prone to socially desirable responses. Another limitation is the fact that the reported conversations may have been happening with the same individuals in the CHL peer groups. Finally, this study being based in a camp setting, in a city as big as Dar es Salaam, makes it not possible to generalize findings to a rural population or to other cities.